Made available in DSpace on 2014-06-11T19:25:36Z (GMT). No. of bitstreams: 0 Previous issue date: 2010-02-12Bitstream added on 2014-06-13T19:32:57Z : No. of bitstreams: 1 tonon_e_me_botfm.pdf: 621514 bytes, checksum: f42de067077b49bbfe8b27db1ee8aff3 (MD5) / Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES) / Fundação de Amparo à Pesquisa do Estado de São Paulo (FAPESP) / Apesar da fisioterapia respiratória aparentemente beneficiar pacientes sob ventilação mecânica, não há evidências suficientes para sua recomendação. Usando associação das manobras compressão torácica (CT) e hiperinsuflação manual (HM), prévio estudo de nosso grupo identificou significante redução no período de ventilação mecânica (VM), no período de internação e melhora da extensão de lesão pulmonar (Murray) em pacientes sob VM. Contudo, é desconhecido o papel isolado de cada manobra nos benefícios encontrados. Portanto, o objetivo deste estudo foi comparar prospectivamente o efeito isolado e associado das manobras CT e HM no período de internação e de VM em pacientes sob VM. O estudo foi conduzido por 13 meses na UTI (Pronto-Socorro do Hospital das Clínicas, UNESP, Botucatu, SP, Brasil) de um hospital universitário terciário. Foi também avaliada a interferência das manobras nos seguintes parâmetros: índice prognóstico (APACHE-II), Murray, oxigenação (PaO2/FiO2), mecânica respiratória, repercussões hemodinâmicas e saturação periférica de oxigênio (SpO2). A análise estatística utilizou o teste de Goodman para contrastes entre e dentro de populações multinomiais, qui-quadrado, análise de variância e análise de variância para o modelo de medidas repetidas em grupos independentes. Dos 204 pacientes que preencheram os critérios de inclusão e exclusão e foram admitidos no estudo, 20 pacientes foram alocados no grupo CT, 20 no grupo HM e 20 no grupo CT+HM de acordo com o processo de sistematização. Diversas causas levaram à exclusão de alguns pacientes durante o estudo e cada grupo passou a ser constituído por 15 pacientes. O grupo CT recebeu compressão torácica, o grupo HM recebeu hiperinsuflação manual e o grupo CT+HM recebeu a associação de ambas as manobras duas vezes ao dia durante cinco... / There is no evidence to support the recommendation of chest physiotherapy on mechanically ventilated (MV) patients, although this procedure apparently improves those patients. Using association of thoracic compression (TC) and manual hyperinflation (MH), our previous study identified significant reduction in duration of weaning from ventilation, discharge from intensive care unit (ICU) and extent of lung damage index (Murray). However, it is unknown the individual role of each maneuver on those benefits. Therefore, the aim of the study was evaluate the isolated and associated effect of TC and MH on the mechanical ventilation period and length of stay in mechanically ventilated patients. Secondarily, outcomes of interest were the effect of physiotherapy on Murray, severity score and on hemodynamics, gas exchange, and respiratory mechanics. It was conducted at ICU of the Emergency Room (ER) at Hospital das Clínicas of São Paulo State University (UNESP-HC) (Botucatu, SP, Brazil) for 13 consecutive months. The significance of differences between groups was accessed by Goodman test, chi-squared analysis, ANOVA and a nonparametric repeated measures ANOVA. The present study was a three-group (TC, MH, and TC+MH), prospective and systematized clinical study lasting 5 days. Of the 204 patients who fulfilled all the inclusion criteria and were enrolled in the study, 20 patients were allocated into TC group, 20 into MH group or 20 into TC+HM group. The TC group received expiratory chest compression, the MH group received manual hyperinflation and the TC+HM received manual hyperinflation combined with expiratory chest compression twice a day for 5 days. Five patients from TC, 5 from MH and 5 from TC+HM were withdrawn during the study period due to several reasons and therefore, 15 patients remained in each group. The 3 groups... (Complete abstract click electronic access below)

Diversos serviços de saúde no Brasil vem apresentado episódios de superlotação, em um contexto onde os recursos são limitados. A redução do tempo de permanência em internações hospitalares tem como consequência direta a disponibilização de mais leitos-dia. O gerenciamento e melhoria do fluxo de pacientes ao longo das internações hospitalares é importante, sendo que o uso eficiente dos leitos pode acontecer devido a uma série de fatores. Estudos demonstram que equipes multidisciplinares podem realizar uma assistência de qualidade, reduzindo custos e o tempo em que os pacientes permanecem internados, sem impacto na reinternação ou mortalidade. Também há trabalhos que apontam a eficácia de unidades dedicadas ao atendimento de doenças específicas. A admissão de pacientes dentro de critérios bem definidos aumenta o giro de leitos. Com o objetivo de analisar se a equipe multidisciplinar Medicina Interna – Emergência (MIE) poderia contribuir para a redução do tempo de permanência hospitalar dos pacientes portadores de doenças prevalentes, sem alterar os indicadores de reinternação e mortalidade, o presente estudo experimental, controlado, não-randomizado, comparou o período pré e pós-intervenção, ou seja, a criação de uma Unidade de Curta Permanência no Serviço de Medicina Interna, do Hospital de Clínicas de Porto Alegre (HCPA). Foram analisadas internações ocorridas através da Emergência do HCPA, de pacientes com 14 anos ou mais, com as doenças prevalentes classificadas conforme grupos do CID-10 (J09-J018; J40-J47; N30-N39; I30-I52; I60-I69; B20-B24; C15-C26; A30-A49; e E10-E14), no período compreendido entre 01 de dezembro de 2008 a 30 de novembro de 2010 (n = 11040). Os resultados do estudo demonstram que após a criação da equipe E-MEI e a sua unidade de curta permanência, houve uma redução do tempo de permanência dos pacientes internados pelas causas selecionadas (antes: 10,89 ± 13,17 dias, após: 9,47 ± 11,24 dias, p = 0,006), e uma diminuição mais acentuada nas internações do Serviço de Medicina Interna [antes (n = 680): 14,33 ± 14,57 dias, após (n = 1243): 9,77 ± 10,62 dias, p = 0,000]. Não ocorreu alteração na taxa de mortalidade de todos os pacientes admitidos para as causas selecionadas [antes (n = 3800): 11,3%, após (n = 3958): 11,8% p = 0,123]. Também não houve alteração na taxa de reinternação de 7 dias na amostra estudada [antes (n = 3369): 7,2%, depois de (n = 3491): 6,7%, p = 0,407]. / Several health services in Brazil has shown episodes of overcrowding, in a context where resources are limited. Reducing the length of stay in hospital has as a direct consequence the provision of more beds-day. Managing and improving the flow of patients throughout the hospital is important, and the efficient use of beds can happen due to a number of factors. Studies have shown that multidisciplinary teams can perform quality care, reducing costs and the time patients remain hospitalized, with no impact on mortality or rehospitalization. There are also studies that show the effectiveness of units dedicated to the treatment of specific diseases. The admission of patients into well-defined criteria increases the turnover of beds. With the objective of analyzing the multidisciplinary team Internal Medicine – Emergency, could help to reduce the length of hospital stay of patients with diseases prevalent, without changing the indicators of rehospitalization and mortality, the present study experimental, controlled, not -randomized study compared the pre-and post-intervention, ie the creation of a Short Stay Unit in the Department of Internal Medicine, Hospital de Clinicas de Porto Alegre (HCPA). We analyzed hospital admissions through the Emergency HCPA, for patients aged 14 years or older with prevalent disease groups classified according to the ICD-10 (J09-J018, J40-J47, N30-N39, I30-I52, I60-I69; B20-B24, C15-C26, A30-A49, and E10-E14), during the period from December 1, 2008 to November 30, 2010 (n = 11,040). The study results show that after the creation of the multidisciplinary team, and its Short Stay Unit, there was a reduction in the length of stay of inpatients by selected causes (before: 10.89 ± 13.17 days after: 9 47 ± 11.24 days, p = 0.006) and a greater reduction in hospitalizations Service of Internal Medicine [before (n = 680): 14.33 ± 14.57 days after (n = 1243): 9, 77 ± 10.62 days, p = 0.000]. No change in the mortality rate of all patients admitted to selected causes [before (n = 3800): 11.3% after (n = 3958): 11.8% p = 0.123]. There was also no change in the rate of readmission than 7 days in our sample [before (n = 3369): 7.2% after (n = 3491): 6.7%, p = 0.407].

Base teórica: Existem controvérsias quanto à quantidade ideal de calorias que pacientes críticos com insuficiência respiratória aguda devem receber, bem como aos efeitos das estratégias de terapia nutricional hipocalórica versus normocalórica nos desfechos clínicos e de tolerância gastrointestinal. Objetivo: Comparar o efeito de duas estratégias de terapia nutricional enteral (nutrição hipocalórica versus normocalórica) nos desfechos clínicos e na tolerância gastrointestinal de pacientes criticamente doentes em insuficiência respiratória aguda. Bases de dados pesquisadas: MEDLINE, EMBASE, SCOPUS e Cochrane Central Register of Controlled Trials até o período de agosto de 2014. Seleção dos estudos: Ensaios clínicos randomizados que compararam o efeito das estratégias de nutrição hipocalórica versus normocalórica nos desfechos clínicos principais [mortalidade na unidade de terapia intesiva (UTI), tempo de internação na UTI e tempo de ventilação mecânica] e nos sinais e sintomas gastrointestinais (regurgitação, aspiração, vômito, diarreia, constipação, distensão abdominal, elevado volume de resíduo gástrico e uso de agentes prócinéticos). Extração dos dados: Informações sobre a execução e qualidade dos estudos e características dos pacientes e dos desfechos de interesse foram extraídas. As estimativas de risco relativo (RR) e média da diferença (MD) foram sintetizadas sob o modelo de efeitos aleatórios. A heterogeneidade foi avaliada com Teste Q e I2. A análise de sensibilidade foi conduzida através de análise de subgrupos os quais foram classificados conforme a estratégia de terapia nutricional enteral utilizada (nutrição trófica versus nutrição hipocalórica moderada). A metanálise foi realizada com apoio do software RevMan v5.3. Resultados: Dentre os 798 estudos encontrados, quatro ensaios clínicos randomizados que avaliaram 1540 pacientes foram incluídos na avaliação qualitativa e quantitativa. Não houve diferença na mortalidade geral (RR, 0.92; 95% CI, 0.73 – 1,19; I2 31% p=0.23 para heterogeneidade). A análise de subgrupos verificou mortalidade geral significativamente menor no subgrupo que recebeu 59-72% das necessidades nutricionais (RR, 0.72; 95% CI, 0.53 – 0.98; I2 0% p=0.78 para heterogeneidade). Não foram encontradas diferenças entre os grupos quanto à mortalidade na UTI, tempo de permanência na UTI ou hospitalar e tempo de ventilação mecânica. Quanto à avaliação da tolerância gastrointestinal, o grupo que recebeu nutrição hipocalórica foi associado a uma menor ocorrência de vômitos, diarreia e constipação quando comparado ao grupo nutrição normocalórica. Não foram verificadas diferenças entre os grupos quanto aos sintomas de aspiração e distensão abdominal. Conclusão: A estratégia de terapia nutricional enteral hipocalórica em aporte moderado (59- 72%) foi associada à menor mortalidade geral. A tolerância gastrointestinal foi superior no grupo que recebeu nutrição hipocalórica. A oferta de terapia nutricional enteral hipocalórica em aporte moderado deve ser preferida em pacientes criticamente doentes. / Context: Controversy exists regarding the optimal amount of calories that critically ill patients with acute respiratory failure should consume as far as clinical outcomes and gastrointestinal tolerability are concerned. Objective: To compare the effect of two enteral nutrition strategies (underfeeding versus fullfeeding) on clinical outcomes and gastrointestinal tolerability in critically ill patients with acute respiratory failure. Data Sources: MEDLINE, EMBASE, SCOPUS and the Cochrane Central Register of Controlled Trials up to August 2014. Study Selection: Randomized Controlled Trials that compared the effects of underfeeding with full-feeding strategies on major clinical outcomes (ICU and overall mortality, ICU and hospital length of stay and mechanical ventilation) and gastrointestinal signs and symptoms (regurgitation, aspiration, vomiting, diarrhea, constipation, abdominal distention, elevated gastric residual volume and use of prokinetic agents). Data extraction: Studies’ information, patient’s characteristics and outcomes were extracted. Risk ratio (RR) and Mean Difference (MD) estimates were synthesized under a randomeffects model. Heterogeneity was evaluated using the Q test and I2. A sensitivity analysis on overall mortality was conducted, wherein the groups were classified according to the feeding strategy used (trophic versus hypocaloric nutrition). Meta-analyses were performed using RevMan v5.3 analysis software. Data synthesis: Among the 798 studies retrieved, four studies of 1540 patients were included. Interventional studies comparing underfeeding with full-feeding were not associated with significant difference in overall mortality (RR, 0.92; 95% CI, 0.73 – 1,19; I2 31% p=0.23 for heterogeneity). Subgroup analysis of the groups according to the amount of delivered calories showed that the overall mortality was significantly lower in the subgroup that achieved 59-72% of energy intake than in the full-feeding group (RR, 0.72; 95% CI, 0.53 – 0.98; I2 0% p=0.78 for heterogeneity). No differences were found between the underfeeding versus full-feeding groups regarding in the ICU mortality, ICU and hospital length of stay and duration of mechanical ventilation. As far as gastrointestinal tolerability is concerned, the underfeeding group showed lower occurrence of vomiting, regurgitation, use of prokinetic agents, elevated gastric residual volume occurrence, diarrhea and constipation when compared with the full-feeding strategy. No differences between the two groups were found for aspiration and abdominal distention. Conclusion: The underfeeding strategy was associated with lower overall mortality in the subgroup that achieved initial moderate intake. Gastrointestinal tolerability was improved by the underfeeding strategy. Initial moderate intake should be preferred rather than trophic or full-feeding in critically ill patients.

Base teórica: Existem controvérsias quanto à quantidade ideal de calorias que pacientes críticos com insuficiência respiratória aguda devem receber, bem como aos efeitos das estratégias de terapia nutricional hipocalórica versus normocalórica nos desfechos clínicos e de tolerância gastrointestinal. Objetivo: Comparar o efeito de duas estratégias de terapia nutricional enteral (nutrição hipocalórica versus normocalórica) nos desfechos clínicos e na tolerância gastrointestinal de pacientes criticamente doentes em insuficiência respiratória aguda. Bases de dados pesquisadas: MEDLINE, EMBASE, SCOPUS e Cochrane Central Register of Controlled Trials até o período de agosto de 2014. Seleção dos estudos: Ensaios clínicos randomizados que compararam o efeito das estratégias de nutrição hipocalórica versus normocalórica nos desfechos clínicos principais [mortalidade na unidade de terapia intesiva (UTI), tempo de internação na UTI e tempo de ventilação mecânica] e nos sinais e sintomas gastrointestinais (regurgitação, aspiração, vômito, diarreia, constipação, distensão abdominal, elevado volume de resíduo gástrico e uso de agentes prócinéticos). Extração dos dados: Informações sobre a execução e qualidade dos estudos e características dos pacientes e dos desfechos de interesse foram extraídas. As estimativas de risco relativo (RR) e média da diferença (MD) foram sintetizadas sob o modelo de efeitos aleatórios. A heterogeneidade foi avaliada com Teste Q e I2. A análise de sensibilidade foi conduzida através de análise de subgrupos os quais foram classificados conforme a estratégia de terapia nutricional enteral utilizada (nutrição trófica versus nutrição hipocalórica moderada). A metanálise foi realizada com apoio do software RevMan v5.3. Resultados: Dentre os 798 estudos encontrados, quatro ensaios clínicos randomizados que avaliaram 1540 pacientes foram incluídos na avaliação qualitativa e quantitativa. Não houve diferença na mortalidade geral (RR, 0.92; 95% CI, 0.73 – 1,19; I2 31% p=0.23 para heterogeneidade). A análise de subgrupos verificou mortalidade geral significativamente menor no subgrupo que recebeu 59-72% das necessidades nutricionais (RR, 0.72; 95% CI, 0.53 – 0.98; I2 0% p=0.78 para heterogeneidade). Não foram encontradas diferenças entre os grupos quanto à mortalidade na UTI, tempo de permanência na UTI ou hospitalar e tempo de ventilação mecânica. Quanto à avaliação da tolerância gastrointestinal, o grupo que recebeu nutrição hipocalórica foi associado a uma menor ocorrência de vômitos, diarreia e constipação quando comparado ao grupo nutrição normocalórica. Não foram verificadas diferenças entre os grupos quanto aos sintomas de aspiração e distensão abdominal. Conclusão: A estratégia de terapia nutricional enteral hipocalórica em aporte moderado (59- 72%) foi associada à menor mortalidade geral. A tolerância gastrointestinal foi superior no grupo que recebeu nutrição hipocalórica. A oferta de terapia nutricional enteral hipocalórica em aporte moderado deve ser preferida em pacientes criticamente doentes. / Context: Controversy exists regarding the optimal amount of calories that critically ill patients with acute respiratory failure should consume as far as clinical outcomes and gastrointestinal tolerability are concerned. Objective: To compare the effect of two enteral nutrition strategies (underfeeding versus fullfeeding) on clinical outcomes and gastrointestinal tolerability in critically ill patients with acute respiratory failure. Data Sources: MEDLINE, EMBASE, SCOPUS and the Cochrane Central Register of Controlled Trials up to August 2014. Study Selection: Randomized Controlled Trials that compared the effects of underfeeding with full-feeding strategies on major clinical outcomes (ICU and overall mortality, ICU and hospital length of stay and mechanical ventilation) and gastrointestinal signs and symptoms (regurgitation, aspiration, vomiting, diarrhea, constipation, abdominal distention, elevated gastric residual volume and use of prokinetic agents). Data extraction: Studies’ information, patient’s characteristics and outcomes were extracted. Risk ratio (RR) and Mean Difference (MD) estimates were synthesized under a randomeffects model. Heterogeneity was evaluated using the Q test and I2. A sensitivity analysis on overall mortality was conducted, wherein the groups were classified according to the feeding strategy used (trophic versus hypocaloric nutrition). Meta-analyses were performed using RevMan v5.3 analysis software. Data synthesis: Among the 798 studies retrieved, four studies of 1540 patients were included. Interventional studies comparing underfeeding with full-feeding were not associated with significant difference in overall mortality (RR, 0.92; 95% CI, 0.73 – 1,19; I2 31% p=0.23 for heterogeneity). Subgroup analysis of the groups according to the amount of delivered calories showed that the overall mortality was significantly lower in the subgroup that achieved 59-72% of energy intake than in the full-feeding group (RR, 0.72; 95% CI, 0.53 – 0.98; I2 0% p=0.78 for heterogeneity). No differences were found between the underfeeding versus full-feeding groups regarding in the ICU mortality, ICU and hospital length of stay and duration of mechanical ventilation. As far as gastrointestinal tolerability is concerned, the underfeeding group showed lower occurrence of vomiting, regurgitation, use of prokinetic agents, elevated gastric residual volume occurrence, diarrhea and constipation when compared with the full-feeding strategy. No differences between the two groups were found for aspiration and abdominal distention. Conclusion: The underfeeding strategy was associated with lower overall mortality in the subgroup that achieved initial moderate intake. Gastrointestinal tolerability was improved by the underfeeding strategy. Initial moderate intake should be preferred rather than trophic or full-feeding in critically ill patients.

INTRODUÇÃO: O diagnóstico do estado nutricional por rastreamento e avaliação nutricional permite detectar desnutrição e se associa com desfechos clínicos negativos em pacientes adultos hospitalizados. OBJETIVO: Identificar o teste mais adequado para avaliação de risco e estado nutricional em relação a desfechos clínicos negativos em pacientes adultos hospitalizados; e investigar a complementaridade existente entre os testes de rastreamento (NRS 2002) e avaliação nutricional (SGA). MÉTODOS: Estudo prospectivo, sequencial, não intervencionista, realizado em 705 pacientes adultos de ambos os sexos, de distintas enfermarias, no Instituto Central do Hospital das Clínicas da Faculdade de Medicina da Universidade de São Paulo. Em até 48 horas da admissão aplicou-se em todos os pacientes quatro testes de rastreamento e avaliação nutricional (NRS 2002: Triagem de Risco Nutricional 2002, MUST: Triagem Universal de Risco Nutricional, MNA-SF: Mini Avaliação Nutricional Reduzida, e SGA: Avaliação Subjetiva Global). Os pacientes foram seguidos até o desfecho final, obtendo-se as intercorrências clínicas de complicações, tempo de internação prolongado e mortalidade. Analisou-se o desempenho de todos os testes por curvas ROC (Receiver Operating Characteristic Curve) e razão de verossimilhança (LR). Verificou-se a complementaridade entre NRS 2002 e SGA por regressão logística e o número necessário de pacientes a avaliar para encontrar um desfecho negativo (NNS). RESULTADOS: NRS 2002 detectou 27,9% (n=197) de risco nutricional, MUST 39,6% (n=279), MNA-SF 73,2% (n=516) e SGA indicou 38,9% de desnutrição moderada e grave (n=274). Os testes NRS 2002 e SGA mostraram melhor desempenho em predizer desfechos clínicos negativos que MUST e MNA-SF pela curva ROC. NRS 2002 apresentou LR positiva maior que os demais testes para todos os desfechos clínicos. Segundo a regressão logística, 13% (IC 10,0- 17,0%) dos doentes podem ter tempo de internação prolongado, 9% (IC 7,0- 12,0%) complicação moderada ou grave e 1% (IC 0,3 - 2,1%) mortalidade. Para tempo de internação prolongado, os pacientes desnutridos por SGA, classe B (SGA B) aumentam esta probabilidade em 1,9 vezes (IC 1,2-3,2 vezes, p=0,008) e SGA classe C (SGA C) em 3,8 vezes (IC 2,0-7,2 vezes, p<0,0001). Para pacientes em risco nutricional por NRS 2002 (NRS+), a probabilidade de complicação moderada e grave aumenta em 1,9 vezes (IC 1,1-3,5 vezes, p=0,03), em 1,9 vezes (IC 1,1-3,4 vezes, p=0,02) para doentes SGA B e em 17,8 vezes (IC 1,4-5,8 vezes, p=0,003) para doentes SGA C. A probabilidade para mortalidade aumenta em 3,9 vezes (IC 1,2- 13,1 vezes, p=0,03) para pacientes NRS+. O NNS calculado para todos os desfechos clínicos negativos em pacientes NRS+ & SGA C (em risco nutricional por NRS 2002 e desnutridos graves pela SGA) foi menor que para os testes isolados. CONCLUSÕES: NRS 2002 é o melhor teste de rastreamento nutricional. A aplicação de SGA em doentes sob risco nutricional por NRS 2002 aumenta a capacidade de predição de desnutrição em relação a desfechos clínicos negativos. / INTRODUCTION: The diagnosis of nutritional status by nutritional screening and assessment tools detects malnutrition and is associated with negative clinical outcomes in adult hospitalized patients. OBJECTIVE: To identify the most appropriate tool for analysis of nutritional risk and malnutrition in relation to adverse clinical outcomes in adult hospitalized patients, and to investigate the complementarity of the nutritional screening (NRS 2002) and nutritional assessment (SGA) tests. METHODS: A prospective, sequential, non-interventional study, conducted in 705 adult patients of both sexes, from different wards in the Hospital das Clínicas da Faculdade de Medicina da Universidade de São Paulo. Within 48 hours of admission, all the patients were submitted to four nutritional screening and assessment tests (NRS 2002: Nutritional Risk Screening 2002, MUST: Malnutrition Universal Screening Tool, MNA-SF: Mini Nutritional Assessment Short Form and SGA: Subjective Global Assessment). Patients were followed until the final outcome, obtaining clinical outcomes of complications, length of hospital stay and death. The performance of the tests was analyzed by the ROC (Receiver Operating Characteristic) curve and likelihood ratio (LR). The complementarity of screening and assessment tools was analyzed by logistic regression, and the number of patients required to screen was obtained by calculating the number needed to screen (NNS). RESULTS: NRS 2002 detected 27.9% (n = 197) of nutritional risk, MUST 39.6% (n = 279), MNA-SF 73.2% (n = 516), and SGA detected moderate or severe malnutrition in 38.9% of the patients (n = 274). NRS 2002 and SGA had a better performance in predicting adverse clinical outcomes than MUST and MNA-SF confirmed by ROC curve. NRS 2002 had higher positive LR compared to the other tests for all the clinical outcomes. According to the logistic regression analysis, 13% (CI 10.0-17.0%) of the patients may have length of hospital stay, 9% (CI 7.0-12.0%) moderate or severe complications and 1% (CI 0.3 - 2.1%) mortality. For length of hospital stay, malnourished patients by SGA, class B (SGA B) increase this probability in 1.9 times (CI 1.2-3.2 times, p = 0.008) and SGA class C (SGA C) in 3.8 times (CI 2.0-7.2 times, p <0.0001). For patients nutritionally at risk by NRS 2002 (NRS +), the probability of moderate and severe complication increase in 1.9 times (CI 1.1-3.5 times, p = 0.03), in 1.9 times (CI 1.1-3.4 times, p = 0.02) for SGA B patients and 17.8 times (CI 1.4-5.8 times, p = 0.003) for SGA C patients. The probability of mortality increase in 3.9 times (CI 1.2-13.1 times, p = 0.03) for NRS+ patients. The NNS calculated for all adverse clinical outcomes in patients NRS+ & SGA C (at nutritional risk by NRS 2002 and severe malnourished by SGA), was lower than for the test separately. CONCLUSIONS: NRS 2002 is the best test for nutritional risk screening. The application of SGA in nutritionally at risk patients by NRS 2002 increases the predictive capacity of malnutrition in relation to adverse clinical outcomes.

Sepse é causa comum de óbito, e vários fatores prognósticos têm sido identificados. Entretanto, é possível que a rota do paciente séptico no hospital também tenha efeito sobre o prognóstico. Nosso objetivo foi verificar se a rota do paciente séptico antes da admissão na UTI tem efeito sobre a letalidade hospitalar. Métodos Foi realizado um estudo de coorte retrospectiva com 489 pacientes com sepse grave ou choque séptico (idade >=18 anos), internados na Unidade de Terapia Intensiva. Analisamos se a rota está associada a mortalidade hospitalar usando modelo de regressão de Cox com variância robusta. Resultados Dos 489 pacientes, 207 (42,3%) foram diagnosticados com sepse na Unidade de Pronto Atendimento (UPA), 185 (37,8%) em unidade de internação clínica ou cirúrgica (Clínica Médica Cirúrgica - CMC), 56 (13,3%) em Unidade Semi-Intensiva (USI) e 32 (6,5%) em Unidade Terapia Intensiva.(UTI). A maioria (56,6%) dos pacientes era do sexo masculino, a idade média foi de 66,3 anos, 39,8% tinham APACHE II de 25 ou mais, e 77,5% tinham o diagnóstico de choque séptico. A letalidade foi 41,9%. Na análise multivariada com ajuste para diversos fatores prognósticos, incluindo tempo de internação hospitalar antes da admissão na UTI, não houve diferença estatisticamente significativa no risco de óbito entre pacientes com sepse grave diagnosticada na UPA ou CMC (risco relativo [RR] 1,36; intervalo de confiança [IC] 95% 1,00 a 1,83). Porém, o risco de óbito hospitalar foi maior nos pacientes em que a sepse grave foi diagnosticada na USI ou UTI (RR 1,64; IC 95% 1,20 a 2,25). Conclusão A mortalidade dos pacientes com sepse grave ou choque séptico atendidos na CMC é similar à de pacientes com sepse diagnosticada na UPA. Entretanto, o risco de óbito hospitalar foi maior nos pacientes que desenvolveram sepse na USI ou UTI / Sepsis is a common cause of death. Several predictors of hospital mortality have been identified. However, it is possible that the route the septic patient takes within the hospital may also affect endpoints. Thus, our main objective was to verify whether the routes of septic patients before being admitted to ICU affect their in-hospital mortality. Methods Retrospective cohort study of 489 patients with severe sepsis or septic shock (age >= 18 years) admitted to the Intensive Care Unit. We analyzed the impact of route on in-hospital mortality using Cox regression with robust variance. Results Of 489 patients, 207 (42.3%) presented with severe sepsis in the ED, 185 (37.8%) were diagnosed with severe sepsis in the ward, 56 (13.3%) in the step down unit and 32 (6.5%) in the ICU. The mortality rate was 41.9%. The mean age was 66.3 years, and 56.6% were men. APACHE II scores were >25 in 39.8% of patients, and 77.5% were diagnosed with septic shock. In the multivariate analysis, with adjustment for several prognostic factors including length of hospital stay before ICU admission, there was no statistically significant difference in the risk of death between patients who had severe sepsis diagnosed in the ED compared to CMC (relative risk [RR] 1,36; IC 95% 1,00 a 1,83). However, the risk of death was increased in patients who had severe sepsis diagnosed in the step-down unit or ICU (RR 1,64; IC 95% 1,20 a 2,25). Conclusion Patients who have severe sepsis or septic shock diagnosed in the CMC have in-hospital mortality similar to those who present with severe sepsis or septic shock in the ED. However, patients who develop severe sepsis in the step-down unit or ICU have higher mortality

Magister Public Health - MPH / Background: Malnutrition accounted high level of childhood morbidity and mortality in Ethiopia including Durame area. Durame area is one of the food insecure districts in Southern region. As a result of high prevlanece of acute malnutrtion, which is 8.3%, Ministry of Health partnering with World Vision Ethiopia started outpatient therapeutic program (OTP) in seven OTP sites to rehablitate severely malnourished children. Reports indicate that number of factors affect the rehabilitation outcome of children with severe acute malnutrtion in OTP programs. However, there are no studies conducted to assess their contribution in the rehablitaiton outcome. Hence, this study will attempt to investigate these factors and assess their public health significance in Durame area. Aim: To assess the factors affecting the rehabilitation outcome of an OTP for children with severe acute malnutrition in Durame area, Southern Ethiopia Method: the study used a descriptive study with an analytical component. Three-hundred and sixty (360) medical records were calculated during sampling and proportional numbers of medical records were sampled from the seven OTP sites. The medical records were reviewed using semi-structured questionnaires from September 1 to September 10, 2008. The data was entered and analyzed using EPI info version 3.3.2 software. Results: three hundred fifty five (98.6%) of the total sample records were reviewed. Three hundred twenty nine (92.7%) children were cured, 11(3.8%) died, 7 (2%) defaulted and 8 (2.3%) were non-cure. Average weight gain on discharge was 3.4gm/kg/day and the mean length of stay was 55.6 days (SD+14 days). More than 60% of children were admitted in three of the seven OTP sites where Demboya OTP sites taking the larger share. Nearly half of the total children (49.8%) were between 6 to 12 months of age and the median age of admission was 13 months. The male to female ratio in the study population was almost equal. Average family size was 6.3 and 58.3% of children came from households with 6 or more family members. Forty two (11.8%) children in the study had twin. The average walking distance to the OTP sites was 62.9 minutes and two hundred fifty six mothers travelled less than an hour. Most of the children (92.1%) were referred from the community and most of the children were admitted with MUAC followed by pitting edema. One hundred seventy four (49%) of the total children were beneficiaries of GFR. On admission two hundred twenty six (63.7%) children were breastfeeding, 257 (72.4%) had no symptoms of sickness and 327 (92.1%) did not have abnormal physical examination findings. More than half (51.5%) of them did not receive any home visit and the larger share of the home visits (37.3%) were made when children got illnesses. One hundred sixteen (32.7%) children in the study had chronic medical conditions during follow up. Fever or hypothermia (0.6%), dehydration (0.8%), anemia (0.6%), skin infection (1.6%) and Plumpy nut refusal (2.0%) were the main abnormal medical findings during follow up. Assessment of the influence of the socio-demographic and biological characteristics on the rehabilitation outcome indicated that the sites, family size, chronic medical conditions, absenteeism, weight loss, presence of fever or hypothermia, dehydration and anemia had significant association with the treatment outcome (p<0.05). Further analysis for significant variables using regression analysis indicated that absenteeism, chronic medical illness, fever or hypothermia and anemia are predictor variables contributing significant information for the prediction of the treatment outcome (p<0.05). Conclusion: The program has high success rate in terms of increasing cure and decreasing death, default and non-cure rates but it did not meet the minimum international recommendations for average length of stay and average weight gain. The study identified the main socio-demographic and biological characteristics of children with SAM and factors that affect the rehabilitation outcome. Children under the age of 24 months were most affected with SAM and no gender variation. Larger proportions of malnourished children were living in families above the average family size, which had significant association with the outcome. OTP sites were accessible for majority children in the program but higher level of absenteeism which significantly associated with the outcome. The study identified socio-demographic and biological factors that influenced the rehabilitation outcome as well as the predictor variables contributing significant information for the prediction of the treatment outcome. It could assist the program implementers to design appropriate public health measures. The achievement in Durame OTP program indicates effectiveness of community based management of SAM and existing potential to integrate in routine health system in resource scarce setting like Durame. Recommendations: to sustain the achievements and improve the growth areas necessary public health measures are prime importance.

The Ottawa Hospital cancels hundreds of elective surgeries every year due to a lack of beds, and has an average weekday occupancy rate above 100%. Our approach to addressing these issues, by way of informing administrators of resource needs, was to model the flow of patients coming and going from the hospital. We used administrative data from the Ottawa Hospital to build a time-series model of emergency department admissions, and studied models that would predict next-day discharge of patients currently taking up hospital beds. In the latter, we considered population-averaged models for groups of patients based on their primary medical condition, as well as subject-specific models. We included the random effects from subject-specific variation to improve on predictive accuracy over the population- averaged approach. The result was a model that provided more realistic probabilities of discharge, and stable predictive accuracy over patient length of stay.

The supervised and semi-supervised machine learning paradigms hinge on the idea that the training data is labeled. The label quality is often brought into question, and problems related to noisy, inaccurate, or missing labels are studied. One of these is an interesting and prevalent problem in the semi-supervised classification area where only some positive labels are known. At the same time, the remaining and often the majority of the available data is unlabeled, i.e., there are no negative examples. Known as Positive-Unlabeled (PU) learning, this problem has been identified with increasing frequency across many disciplines, including but not limited to health science, biology, bioinformatics, geoscience, physics, business, and politics. Also, there are several closely related machine learning problems, such as cost-sensitive learning and mixture proportion estimation. This dissertation explores the PU learning problem from the perspective of density estimation and proposes a new modular method compatible with the relabeling framework that is common in PU learning literature. This approach is compared with two existing algorithms throughout the manuscript, one from a seminal work by Elkan and Noto and a current state-of-the-art algorithm by Ivanov. Furthermore, this thesis identifies two machine learning application domains that can benefit from PU learning approaches, which were not previously seen that way: predicting length of stay in hospitals and automatic music tagging. Experimental results with multiple synthetic and real-world datasets from different application domains validate the proposed approach. Accurately predicting the in-hospital length of stay (LOS) at the time of admission can positively impact healthcare metrics, particularly in novel response scenarios such as the Covid-19 pandemic. During the regular steady-state operation, traditional classification algorithms can be used for this purpose to inform planning and resource management. However, when there are sudden changes to the admission and patient statistics, such as during the onset of a pandemic, these approaches break down because reliable training data becomes available only gradually over time. This thesis demonstrates the effectiveness of PU learning approaches in such situations through experiments by simulating the positive-unlabeled scenario using two fully-labeled publicly available LOS datasets. Music auto-tagging systems are typically trained using tag labels provided by human listeners. In many cases, this labeling is weak, which means that the provided tags are valid for the associated tracks, but there can be tracks for which a tag would be valid but not present. This situation is analogous to PU learning with the additional complication of being a multi-label scenario. Experimental results on publicly available music datasets with tags representing three different labeling paradigms demonstrate the effectiveness of PU learning techniques in recovering the missing labels and improving auto-tagger performance. / Graduate

Influenza is a highly contagious respiratory virus that can cause very severe health complications in people, and can be especially dangerous for young children. The peak influenza season occurs in the winter months with February usually being the month with the highest number of reported infections. As the virus can cause serious illness, pediatric institutions during the winter months see a very large number of patients who have influenza or influenza related complications. Pediatric Emergency Departments (ED) similarly see a dramatic increase in the number of patients who visit the ED during the winter influenza season. Therefore, it is important that pediatric EDs develop ways to handle the increased patient population while still maintaining quality care to the rest of the ED. Thus, a novel influenza clinic run entirely by non-ED Nurse Practitioners (NP) was implemented into the operations of the ED as an annexed clinic in February 2013 during the winter influenza season. The clinic was beneficial in improving the average quality measures of the ED against similar days without the influenza clinic, lowering the average length of stay (LOS) by 24 minutes (13% decrease) and lowering the left without being seen rates (LWBS) by 1.35% (3 fewer patients on average). In addition, using NPs instead of higher cost physicians dramatically lowered the cost of the clinic by nearly half. While the influenza clinic was beneficial in lowering the average LOS and LWBS rates against similar days without the clinic the data did not reach statistical significance, perhaps due to the small amount of data available. The results, despite the statistical insignificance, show a promising future in addition of an NP run influenza clinic to handle the increased patient population during the winter influenza season.